Activity 2

Load Required Libraries

library(tidyverse)

Registered S3 methods overwritten by 'dbplyr':
  method         from
  print.tbl_lazy     
  print.tbl_sql      
── Attaching packages ──────────────────────────────────────────────────────────────────────────────── tidyverse 1.3.1 ──
✓ ggplot2 3.3.3     ✓ purrr   0.3.4
✓ tibble  3.1.2     ✓ dplyr   1.0.6
✓ tidyr   1.1.3     ✓ stringr 1.4.0
✓ readr   1.4.0     ✓ forcats 0.5.1
── Conflicts ─────────────────────────────────────────────────────────────────────────────────── tidyverse_conflicts() ──
x dplyr::filter() masks stats::filter()
x dplyr::lag()    masks stats::lag()

library(gapminder)

Nest the DataFrame by country and continent

gapminder.nested <- gapminder %>%
  group_by(country, continent) %>%
  nest()


gapminder.nested

Fit a linear regression of gdpPercap on year.

gapminder.lm <- lm(gdpPercap ~ year, data=gapminder)
summary(gapminder.lm)

Call:
lm(formula = gdpPercap ~ year, data = gapminder)

Residuals:
   Min     1Q Median     3Q    Max 
-10507  -5434  -2809   2071 109228 

Coefficients:
              Estimate Std. Error t value Pr(>|t|)    
(Intercept) -249693.71   26678.48  -9.359   <2e-16 ***
year            129.78      13.48   9.630   <2e-16 ***
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 9602 on 1702 degrees of freedom
Multiple R-squared:  0.05167,   Adjusted R-squared:  0.05112 
F-statistic: 92.74 on 1 and 1702 DF,  p-value: < 2.2e-16

Fit a regression for each country in your nested data.frame

country_model <- function(df) {
  lm(gdpPercap ~ year, data = df)
}

gapminder.model  <- gapminder.nested %>% 
  mutate(model = map(data, country_model), 
         coef = map(model, broom::tidy))

Calculate measures for checking model performance

gapminder.coef  <- gapminder.model %>% unnest(coef)
gapminder.coef

gapminder.coef %>% 
  mutate(term = fct_recode(term, 
                           Intercept = "(Intercept)", 
                           Slope = "year")) %>% 
  ggplot(aes(estimate, fill = term)) + 
  geom_density(show.legend = FALSE, alpha = 0.5) + 
  geom_histogram(col = "black", fill = "lightgrey",
                 alpha = 0.5,
                 aes(y = ..density..)) + 
  facet_wrap(~term, scales = "free") + 
  scale_fill_brewer(palette = "Set1") + 
  theme_minimal() + 
  labs(y = NULL, x = "Estimate")

glance_df <- gapminder.model %>%
  mutate(glance = map(model, broom::glance)) %>%
  unnest(glance, .drop = TRUE)

The `.drop` argument of `unnest()` is deprecated as of tidyr 1.0.0.
All list-columns are now preserved.

glance_df

Questions

Question 1

gapminder.model$model[[1]]

Call:
lm(formula = gdpPercap ~ year, data = df)

Coefficients:
(Intercept)         year  
  1674.8134      -0.4406  

glance_df %>%
  select(country, continent, r.squared) %>%
  group_by(continent) %>%
  summarize(r.squared=mean(r.squared)) %>%
  arrange(r.squared)

Question 2

glance_df %>%
  select(country, continent, r.squared) %>%
  filter(r.squared < 0.25) %>%
  arrange(country)

Question 3

glance_df %>%
  select(country, continent, r.squared) %>%
  group_by(continent) %>%
  summarize(r.squared=min(r.squared)) %>%
  arrange(r.squared)

Question 4

glance_df %>%
  select(country, continent, r.squared) %>%
  filter(country == 'Australia')

Question 5

glance_df %>% 
  ggplot(aes(continent, r.squared)) + 
  geom_jitter(aes(color=continent))